What This Checks
tm infer doctor is a local readiness check for the next inference request. It helps you confirm that the CLI has the inputs it needs, but it does not send a live inference request.
Use tm infer chat for an end-to-end gateway check, and use tm auth whoami if you also need to confirm the current Control Plane bearer token.
Usage
Examples
Inspect On-Demand readiness for a direct request and for --model @latest.
Inspect Serverless readiness for a direct request that will provide --model explicitly.
Inspect the effective Serverless target after applying explicit request overrides.
Use shell-friendly exit codes when direct gateway prerequisites are missing.
Options
| Name | Type | Required | Default | Details | |
|---|---|---|---|---|---|
--surface | `choice[on-demand | serverless]` | no | "on-demand" | Inference surface to inspect. |
--model | text | no | Model name to inspect for the next request. | ||
--user-id | text | no | X-User-Id header to inspect. Only used for —surface on-demand. | ||
--api-key | text | no | Inference API key to inspect. | ||
--base-url | text | no | Override the base URL for the selected surface. | ||
--exit-status | boolean | no | false | Exit non-zero when checks report a not-ready state. Boolean flag. |
Inherited Global Options
| Name | Type | Required | Default | Details | ||||
|---|---|---|---|---|---|---|---|---|
--version, -V | boolean | no | false | Show the version and exit. Boolean flag. | ||||
--config | path | no | "~/.config/tensormesh/config.toml" | Path to config TOML file | ||||
--output | `choice[text | json | yaml | raw | table]` | no | "text" | Output format (text is human-readable; json is machine-friendly). |
--quiet | boolean | no | false | Suppress non-essential output. Boolean flag. | ||||
--debug | boolean | no | false | Print debug logs to stderr (secrets redacted). Boolean flag. | ||||
--ca-bundle | path | no | Path to a PEM CA bundle for TLS verification (overrides TENSORMESH_CA_BUNDLE). | |||||
--timeout | float | no | Default HTTP timeout in seconds (overrides TENSORMESH_TIMEOUT_SECONDS; subcommands may override). | |||||
--max-retries | integer | no | Max retries for idempotent HTTP requests on transient errors (overrides TENSORMESH_MAX_RETRIES; subcommands may override). | |||||
--controlplane-base | text | no | Override the Control Plane base URL. | |||||
--gateway-provider | text | no | Inference Gateway provider for built-in host selection (nebius, lambda, yotta). |
Prerequisites
- For the default On-Demand flow, run
tm init --syncfirst so the CLI can reuse the synced managed gateway settings.
Caveats
- Reports local readiness only. Use
tm auth whoamito live-validate Control Plane auth andtm infer chatfor an end-to-end inference check. tm infer doctoraccepts the same local targeting overrides astm infer chatfor--model,--user-id,--api-key, and--base-url.--surface on-demandchecks local prerequisites for X-User-Id and--model @latestresolution.--surface serverlesschecks the shared inference API key (gateway_api_key) and selected host, but does not live-validate the upstream endpoint.- Use
tm billing pricing serverless listto discover published serverless model names when you have Control Plane access for the same Tensormesh environment.tm infer doctorcan check whether--modelis present, but it does not enumerate them itself. --model @latestreadiness only applies to--surface on-demandand only checks for a local Control Plane token; it does not live-validate that token.